What This Looks Like
The user asks for a format, but the instruction is too weak to reliably govern the output. The AI may follow the format partially, add extra text, omit required parts, change labels, reorder fields, or treat the requested structure as a loose preference instead of a contract.
Why It Matters
Weak format rules create unstable outputs. A human may still understand the answer, but a parser, reviewer, template, workflow, or downstream system may not be able to rely on it. This can lead to repeated prompt fixes instead of a stable output contract.
Structural Signal
A formatting constraint exists, but it does not fully declare what the output must include, exclude, preserve, or forbid. The issue is not merely noncompliance; it is that the rule itself does not provide enough structure to enforce the desired form.
Common Triggers
- The prompt says “use this format” without defining required fields
- Examples are treated as suggestions instead of constraints
- The output rule does not forbid commentary or extra sections
- Required ordering, labels, or field behavior is not declared
- The format depends on implied expectations from prior context
- The user expects strict structure but gives natural-language guidance
When to Use This Issue
Use this Issue when the output format fails because the rule governing it is too vague, incomplete, optional, or under-specified to reliably produce the expected structure.
When Not to Use This Issue
Do not use this Issue when the format rule is clear and the AI simply violates it. Do not use it when the issue is a specific missing field, wrong type, or parser failure unless the root problem is the weak rule.